© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 105165
    Prev
    Next
    INDEX
    Explanations

    via

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on the word “via” (and similar source-attribution markers), signaling when the text is crediting or citing a source.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     hvem
    -1.81
     preparada
    -1.76
     hvorfor
    -1.69
     utilizada
    -1.68
     hvordan
    -1.65
    袛
    -1.63
    ).
    -1.59
     größ
    -1.55
     acoged
    -1.55
     hábit
    -1.54
    POSITIVE LOGITS
    떳
    1.93
     via
    1.80
     dibanding
    1.77
    From
    1.68
    様々な
    1.65
    onski
    1.63
     berdua
    1.63
    店舗
    1.60
     perdana
    1.59
    ؟
    1.58
    Activations Density 0.057%

    No Known Activations